摘要
在纯净语音条件下,提出了一种基于短时傅里叶变换的敏感语音提取,该提取利用3种FIR数字滤波器对纯净语音在不同频率范围的特征语音进行选取,从而产生不同的音效。待处理的声音文件可以是自己录制的一段语音,也可以是任意格式的音频文件,通过DSP和编译软件实现敏感语音提取,并在数字处理芯片TMS320C5402上实现。
On condition of pure sound, this paper put forward a sensitive speech extraction based on short-time Fourier transform, which used three FIR digital filter extract sensitive speech in different frequency range. The sound that will be processed can be recorded or other sound with any format, then the extraction is compiled by CCS and realized by digital processing chip TMS320C5402.
出处
《微型机与应用》
2009年第24期78-81,共4页
Microcomputer & Its Applications
